Wandering Thoughts archives


A thesis: noisy email addresses are dead

Here is a thesis, or at least a theory: any email address that gets too much noise (spam, bounces, autoresponder replies, and so on) is effectively dead. This is especially so if the noise reaches real people and must be filtered by hand.

(This thesis underlies my belief that postmaster is a dead address.)

The problem with hand filtering is twofold. First, hand filtering is an unpleasant task (and a pain), which means that it will likely be done reluctantly and as little as possible. Given that it's essentially scutwork, it's also likely to be a morale hit for the people who have to do it; the more spam and the less actual email to be fished out of the sewer, the worse the work is. (Imagine the feeling when there is no real email at all, since you just spent all of the time for essentially nothing. Very few people like to actually waste their time.)

Second, hand filtering is error-prone (especially given the previous). I believe that this has been experimentally demonstrated, and should not really be surprising; humans are prone to snap judgements, skimming, and becoming habituated into seeing what they expect to see. If you deal with a constant stream of spam, your mind is rapidly going to assume that every email you see is spam and automatically hit the 'is spam' button unless there is something about it that catches your eye very fast.

The my concerns with automatic filtering are more indirect and less certain. First off, automatic filtering is not completely error free, which means that you lose a certain amount of real email into the spam bin (and that you deal with a certain amount of spam that gets through to real people, with the above problems). Then people perceive this as flakyness in the address, which pushes them away from using it; your address gets less useful to people and you get less signal in the spam-noise, making the problem worse.

spam/NoisyAddressesAreDead written at 00:43:57; Add Comment

Page tools: See As Normal.
Login: Password:
Atom Syndication: Recent Pages, Recent Comments.

This dinky wiki is brought to you by the Insane Hackers Guild, Python sub-branch.